Your culture problem isn't the trait you picked. It's the one wire it's plugged into.

You have probably tried this: you decide the place has gotten too bureaucratic, or not collaborative enough, or too soft on accountability, and you address it directly — a new value on the wall, a line in the all-hands, a policy tweak. Six months later the survey scores on that one trait have barely moved. A new paper suggests why: you may have been turning a dial that isn't wired to anything.

The paper is by Senne Letouche, Sofie Dupré, and Bart Wille of Ghent University and Maarten De Schryver of Poolstok and Ghent University, published September 15, 2026 in Frontiers in Organizational Psychology. I read the full text directly; it is open access. Letouche is also employed by Securex and De Schryver by Poolstok, both Belgian HR services firms — worth knowing, since that's likely how the authors got real internal survey data rather than a recruited sample.

Most culture surveys, including the well-known Competing Values Framework used here, score an organization on broad dimensions — clan, adhocracy, market, hierarchy — by averaging groups of items together. The authors argue this throws away information: it assumes every item in a dimension matters equally and that dimensions don't talk to each other. Instead, using data from a Belgian public-service organization (4,270 employees, 98% response rate) and a Belgian healthcare organization (338 employees), they modeled all 16 individual culture items as a network — a web where each item is a node and its statistical connection to every other item is an edge.

Two things came out of that web worth sitting with.

First: not every node matters equally, and the most influential one wasn't what you'd guess. Of all 16 items, the one most tightly wired into perceptions of everything else was whether managers were seen as "coordinators and coaches who help employees meet the facility's goals and objectives." Not the innovation items. Not the family-and-loyalty items. Goal-setting and coaching.

Results suggest that the degree to which leaders formulate challenging objectives is highly influential for employees' perceptions of the organizational culture.Letouche, Dupré, De Schryver & Wille, Frontiers in Organizational Psychology, 2026 — read in the full published article

The paper also found, in both organizations independently, that leaders seen as rule-enforcers were tied to perceptions of a bureaucratic environment, and that warm, mentoring manager behavior was tied to lower perceptions of a cutthroat culture. Not shocking alone. What's worth noticing: it only shows up at the individual-behavior level. Average it into a broad "hierarchy" or "market" score, as most culture surveys do, and the specific lever disappears into the summary number.

Second, and this cuts against the study's own hypothesis: the authors expected employees with over 20 years at the public-service organization (n = 1,349) to have a more tightly locked-in, harder-to-shift picture of the culture than employees under 5 years (n = 1,002) — the standard assumption about long tenure. Tested directly, the two groups' networks were not significantly more or less interconnected overall. What differed was which specific beliefs were fused together — for veterans, "leaders enforce rules" was bound tighter to "this is a stable place"; for newer employees, to something closer to a family-loyalty read of that same behavior. Same signal, different wiring, no difference in overall rigidity.

The authors are explicit about where this doesn't reach. In their own words, the two organizations were "public service and health care firms, and therefore not the typical for-profit firms" the survey instrument was built around, and "caution is warranted when generalizing the specific network structures reported here." They also flag the data as cross-sectional, so which behavior drives which perception, rather than just traveling with it, remains, in their word, "putative." This is a demonstration of a method on two specific organizations, not a verdict on what drives culture at a company like yours.

Here is why the method matters more to you than to a large company's HR department. In a business like yours, there's no layer of middle management diluting the signal — the specific way you set goals, or your GM enforces the two rules he actually enforces, is a much larger share of the total wiring than in an organization with thousands of employees. If one behavior is doing outsized work shaping how your culture reads, it's more likely to be a behavior of yours specifically.

The part the research doesn't say, but that I'll say plainly: you don't need a network model to use this. Stop assuming your culture is a set of independent dials, and ask your people which one behavior, if it changed, would change how everything else feels. Ask eight or ten of them, separately. If the same one or two answers keep surfacing — probably about how expectations get set, not the poster in the break room — that's your central node, and it's likely the only lever worth pulling first.

And don't assume your longest-tenured people are the wall. This study didn't find that tenure predicts how locked-in someone's view of the culture is — only that it predicts which things are locked together. The colleague who's been there fifteen years may not be more resistant to your next change. They may just be running the same signal through a different wire.

If you want a clearer read on which of your own behaviors are doing that kind of outsized, wiring-level work, the Leadership Profile is built to surface exactly that.